Short answer: invest in custom software when a process is important to how you operate or compete, when the tools you can buy force expensive workarounds, and when the cost of the current way of working is clearly higher than the cost of building and maintaining something better. If a product you can buy or configure already fits, use that instead.
Seven signs it is time
- A critical process runs on spreadsheets. Payroll inputs, leave, stock, client records or compliance tracking live in files that only one or two people understand.
- The same data is typed in more than once. Staff re-capture information from email into a spreadsheet and then into another system.
- Managers cannot get a current view. Reports are compiled by hand, are out of date by the time they arrive, and differ depending on who made them.
- Your process is a genuine differentiator. The way you schedule, quote, deliver or serve clients is part of why customers choose you, and generic software would flatten it.
- You pay for several tools that don't talk to each other. Licence costs and manual hand-offs add up.
- Growth multiplies the admin. Every new branch, site or client adds a proportional amount of manual work.
- Compliance risk is rising. Personal information sits in shared drives and inboxes with no access control or audit trail, which is a POPIA concern. Our POPIA checklist for business software covers what to look for.
If several of these sound familiar, read why South African businesses should stop running critical processes on spreadsheets.
When custom software is the wrong answer
- The process is standard. Accounting, payroll and email are well served by established products.
- The process itself is unclear. Software cannot fix a process that the business has not agreed on. Map it first.
- There is no one to own it. Custom software needs a business owner who makes decisions, tests, and drives adoption.
Often the right answer sits in between: buy the standard parts, configure what you already own, and build only what is specific to your business. Our guide to whether to build, buy or configure explains how to decide.
A simple test
Estimate the hours a process takes each week, multiply by the people involved and their hourly cost, and annualise it. Compare that with the cost of building and maintaining a system over a few years. Include the less visible costs too: errors, delays, staff frustration and risk. If the gap is clear, it is worth a scoped conversation. Our guide to what custom software costs in South Africa sets out the price bands and what moves a project within them.
How to start small
- Pick one process with a clear owner and a visible cost.
- Do a short discovery to define scope.
- Build the smallest version that removes the most manual work.
- Use it, measure it, then decide on the next phase.
Example
An organisation handling leave through forms, email approvals and spreadsheets moved to a structured workflow: employee request, manager review, HR administration, with balances and reports in one place. See the CIT Leave Management System case study.
Key takeaways
- Invest when the process matters, the workarounds are costly and the gap is clear.
- Don't build for standard processes, unclear processes or processes nobody owns.
- Compare the yearly cost of the current way of working with the cost of building and maintaining a system.
- Start with one process, deliver in phases and measure as you go.
Custom software is worth it when it removes a cost the business is already paying every week, not when it simply replaces one tool with another.
Next step: if you are unsure whether to build, buy or configure, start with technology consulting, or read about custom software development.
